Abstract Clear cell renal cell carcinoma (ccRCC) is a highly vascularized and immune infiltrated cancer, and is responsible for the majority of deaths caused by kidney cancer. Notably, ccRCC is defined by clear mutational events, starting with VHL loss, and followed by loss of an epigenetic regulator (e.g. BAP1, PBRM1) on chromosome 3p. While clinical studies have associated BAP1 driven tumors with worse prognosis relative to PBRM1 driven cases, at present histological grade is the only feature distinguishing these subtypes. Moreover, attempts to study the gene expression characteristics of renal cell carcinoma have been limited to tumor cells, leaving the precise role of the immune compartment in disease progression uncharacterized, especially in relation to genetic driver subtype. Here, I present a set of complementary approaches to examine the relationship between genetic subtype and phenotype in ccRCC at the tissue and molecular levels. In Aim 1, I will develop a model to create condensed representations of ccRCC slide images to identify underlying spatial configurations of tissue, and how they vary across BAP1 and PBRM1 driven tumors. In Aim 2, I will focus on the immune compartment ? the infiltrating and bordering immune populations of the tumor, and will identify patterns of gene expression associated with both intrinsic and acquired resistance to immunotherapy. This work stands to provide clarity on the relationship between driving genetic events and fundamental renal cell carcinoma biology, and consequently has the potential to immediately impact clinical decision making.